[
https://issues.apache.org/jira/browse/DIGESTER-163?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13231548#comment-13231548
]
Thomas Neidhart edited comment on DIGESTER-163 at 3/16/12 7:34 PM:
-------------------------------------------------------------------
I have no idea about the digester, but I looked at this issue out of curiosity
(and I like to debug concurrency problems ;-).
The problems is in the DigesterLoader#addRules method:
* createRuleSet clears the underlying rule binder
* addRuleInstance iterates over the same rule binder
If you make this method synchronized, you will not get an exception anymore.
What I do not know, if the createRuleSet is implemented correctly. I would
expect that the initialization of the rules binder happens when the
DigesterLoader is created, not when a new digester is requested, but this is
may be due to lazy loading?
was (Author: tn):
I have no idea about the digester, but I looked at this issue out of
curiosity (and I like to debug concurrency problems ;-).
The problems is in the DigesterLoader#addRules method:
* createRuleSet clears the underlying rule binder
* addRuleInstance iterates of the same rule binder
If you make this method synchronized, you will not get an exception anymore.
What I do not know, if the createRuleSet is implemented correctly. I would
expect that the initialization of the rules binder happens when the
DigesterLoader is created, not when a new digester is requested, but this is
may be due to lazy loading?
> ConcurrentModificationException creating a new Digester via
> loaderInstance.newDigester()
> ----------------------------------------------------------------------------------------
>
> Key: DIGESTER-163
> URL: https://issues.apache.org/jira/browse/DIGESTER-163
> Project: Commons Digester
> Issue Type: Bug
> Affects Versions: 3.2
> Environment: Linux, JDK 6
> Reporter: Torsten Krah
> Attachments: 163-2.patch, 163.patch, Digester163TestCase.java,
> cli-mvn-test-withfix.txt, stack-afterfix.txt, stack-mvn.txt, stack-next.txt,
> stack-next2.txt
>
>
> I am gettig a ConcurrentModificationException when trying to create new
> Digester instance from a configured loader:
> Trace is:
> {code}
> java.util.ConcurrentModificationException: null
> at
> java.util.LinkedList$ListItr.checkForComodification(LinkedList.java:761)
> ~[na:1.6.0_27]
> at java.util.LinkedList$ListItr.next(LinkedList.java:696) ~[na:1.6.0_27]
> at
> org.apache.commons.digester3.binder.FromBinderRuleSet.addRuleInstances(FromBinderRuleSet.java:130)
> ~[commons-digester3-3.2.jar:3.2]
> at
> org.apache.commons.digester3.binder.DigesterLoader.addRules(DigesterLoader.java:581)
> ~[commons-digester3-3.2.jar:3.2]
> at
> org.apache.commons.digester3.binder.DigesterLoader.newDigester(DigesterLoader.java:568)
> ~[commons-digester3-3.2.jar:3.2]
> at
> org.apache.commons.digester3.binder.DigesterLoader.newDigester(DigesterLoader.java:516)
> ~[commons-digester3-3.2.jar:3.2]
> at
> org.apache.commons.digester3.binder.DigesterLoader.newDigester(DigesterLoader.java:475)
> ~[commons-digester3-3.2.jar:3.2]
> at
> org.apache.commons.digester3.binder.DigesterLoader.newDigester(DigesterLoader.java:462)
> ~[commons-digester3-3.2.jar:3.2]
> {code}
> The binder documentation (employee servlet) and the mailing list did confirm
> to me, that the loader should be safe to be shared, so this should not happen.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
